NVIDIA/SkillEvaluator
598 75PythonApache-2.0Updated 2 weeks ago
Multi-tier framework for evaluating AI agent skills with quality gates, semantic overlap detection, synthetic evaluation dataset generation, and live agent evaluation that measures how skills affect agent behavior.
agent-evaluationagent-securityagent-skillsagentic-aibenchmarkclaude-codecodexevaluateevaluationsecurity-scannerskill-evalskill-evalsskill-evaluationskill-evaluatorskills
Our verdict: worth keeping an eye on
A comprehensive Python framework for measuring AI‑agent skill impact; useful for benchmarking Claude‑Code hooks but heavyweight and not immediately required
Filed under dev tooling in our directory.